Retrieval-Induced Upregulation of Tet3 in Pyramidal Neurons of the Dorsal Hippocampus Mediates Cocaine-Associated Memory Reconsolidation

Retrieval-Induced Upregulation of Tet3 in Pyramidal Neurons of the Dorsal Hippocampus Mediates... Background: Memory retrieval refers to reexposure to information previously encoded and stored in the brain. Following retrieval, a once-consolidated memory destabilizes and undergoes reconsolidation, during which gene expression changes to restabilize memory. Investigating epigenetic regulation during reconsolidation could provide insights into normal memory formation and pathological memory associated with psychiatric disorders. Methods: We used cocaine-induced conditioned place preference to assess the cocaine-associated memory of mice and used chemogenetic methods to manipulate the activity of the pyramidal neurons in the dorsal hippocampus. We isolated the ribosome-associated transcripts from the excitatory neurons in the dorsal hippocampus by RiboTag purification to identify the potential epigenetic regulators, and we specifically knocked down gene expression in pyramidal neurons with a Cre- dependent lentivirus. Results: Chemogenetically silencing the activity of the pyramidal neurons in the dorsal hippocampus immediately after memory retrieval markedly impaired memory reconsolidation, and the ribosome-associated mRNA level of the ten-eleven translocation (Tet) family methylcytosine dioxygenase Tet3, but not Tet1 or Tet2, was dramatically upregulated 10 minutes after memory retrieval. The protein level of Tet3 in the dorsal hippocampus but not in the anterior cingulate cortex was dramatically increased 1 hour after memory retrieval. Specifically, knockdown of Tet3 in pyramidal neurons in the dorsal hippocampus decreased the activation of pyramidal neurons and impaired the reconsolidation of cocaine-associated memory. Conclusions: Our findings highlight the new function of the DNA demethylation regulator Tet3 in pyramidal neurons of the dorsal hippocampus in regulating the reconsolidation of cocaine-associated memory. Keywords: reconsolidation, dorsal hippocampus, CamkIIα neuron, Tet3, cocaine-associated memory Introduction The formation of long-term memory involves a series of protein synthesis, and neuronal plasticity dynamics (Flavell molecular and signaling changes, including gene transcription, et al., 2013). These changes may occur during learning and can Received: August 8, 2017; Revised: October 6, 2017; Accepted: October 27, 2017 © The Author(s) 2017. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, 255 provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 256 | International Journal of Neuropsychopharmacology, 2018 Significance Statement Immediately after memory retrieval, a once-consolidated memory destabilizes and undergoes reconsolidation, during which gene expression changes to restabilize memory. Investigating the epigenetic regulators that act during reconsolidation could provide insights into normal memory formation and pathological memory associated with psychiatric disorders. We found that the activity of the excitatory pyramidal neurons in the dorsal hippocampus (dHC) was required for cocaine-associated memory reconsolidation. After memory retrieval, the number of 5hmC pyramidal neurons and the ribosome-associated transcripts of the ten-eleven translocation (Tet) family methylcytosine dioxygenase Tet3, but not Tet1 or Tet2, were upregulated in the dHC. The protein level of Tet3 in the dHC was also increased after memory retrieval. Specifically, knockdown of Tet3 in pyramidal neurons of the dHC impaired the activation of pyramidal neurons and memory reconsolidation. Our findings indicate the importance of DNA demethylation of dHC in regulating reward memory reconsolidation and reveal Tet3 as a new regulator of reward memory. be subsequently retained for a long time. Memory reconsoli- Tet3 may represent a potential treatment target for memory- dation is the process in which previously consolidated memo- related disorders, including drug addiction. ries, having been reactivated, are stored again to maintain, strengthen, or modify existing memories (Haubrich and Nader, METHODS 2016). Following retrieval, once-consolidated memory destabi- lizes and requires gene expression to be restabilized (Jarome Animals and Housing and Lubin, 2014). The result of retrieval is a cascade of events T29-1Stl/J that ultimately leads to changes in neuronal plasticity, which CamkIIα-Cre (#005359 B6.Cg-Tg(Camk2a-cre) ), RPL22-HA tm1.1Psam/J mediate long-term memory updating (Alberini and Kandel, (#011029-B6N.129-Rpl22 ) mice were purchased from the 2014). Thus, understanding the molecular mechanisms of gene Jackson Laboratory. All mice were bred onto a C57BL/6J genetic transcription regulation during the reconsolidation processes background. We generated the CamkIIα-Cre::RPL22-HA (+/-) mice could provide crucial insights into normal memory formation by crossing CamkIIα-Cre mice with RPL22-HA mice. The 8- to and pathological memory associated with psychiatric disorders, 12-week-old male offspring were used in the experiments (see such as PTSD and addiction (Jarome and Lubin, 2014). Figure  3a for a breeding scheme). Genotypes were determined Memory retrieval induces de novo transcription and trans- by polymerase chain reaction (PCR) of mouse tail DNA samples. lation, which are orchestrated by epigenetic modifications and Mice were housed in groups on a 12-hour-light/-dark cycle with are pivotal for memory maintenance and modification (Alberini food and water available ad libitum. All animal treatment was and Kandel, 2014). Epigenetic mechanisms have been widely in strict accordance with the National Institutes of Health Guide implicated in synaptic plasticity underlying memory formation for the Care and Use of Laboratory Animals and was approved (Day and Sweatt, 2010; Sultan and Day, 2011; Zovkic et al., 2013). by the Animal Care and Use Committee of the School of Basic DNA methylation and histone modifications have emerged as Medical Sciences of Fudan University. critical transcriptional regulators of gene expression during ini- tial memory encoding (Jarome and Lubin, 2014), whereas studies Stereotaxic Surgery on epigenetic regulation in memory reconsolidation are few and mainly focused on histone acetylation (Bredy and Barad, 2008; Mice were anesthetized with choral hydrate (40 mg/kg, i.p.) and Maddox et  al., 2013; Graff et  al., 2014). Recent studies indicate placed in a stereotactic instrument. Guide cannulae (Plastics that DNA methylation is involved in the regulation of genomic One) were bilaterally implanted in the brain and affixed to responses during reconsolidation (Flavell et  al., 2013 Oli ; veira, the skull with dental cement. The intended stereotactic coor - 2016). By contrast, it remains unclear how memory retrieval dinates were as follows: dorsal hippocampus – anterior-poste- triggers epigenetic changes, especially demethylation changes, rior (AP) -1.5  mm; medial-lateral (ML) ±1.5  mm; dorsal-ventral during the reconsolidation phases to impact behavior, and new (DV) -1.5 mm; ACC – AP +0.5 mm; ML ±0.5 mm; DV -1.2 mm. All epigenetic regulators of gene transcription during memory re- mice were given at least 10  days to recover before behavioral consolidation remain to be identified. experiments. Anisomycin (A601115-0005, BBI) was dissolved Tet3 belongs to the ten-eleven translocation family of at 125 μg/μL as previously described (Nader et  al., 2000). Each methylcytosine dioxygenases, and it promotes DNA demethyla- mouse received a microinjection of 0.5 μL of anisomycin or vehi- tion (Langemeijer et al., 2009; Gu et al., 2011; Shen et al., 2014). cle through the cannula (0.2  μL/min) immediately after retrieval. Despite the abundance of Tet proteins in the brain, little is known about the dynamic changes and the functions of Tet enzymes Cocaine-Induced CPP after memory retrieval. In this study, by using a cocaine-induced CPP induced by cocaine hydrochloride (Qinghai Pharmaceutical conditioned place preference (CPP) paradigm in mice, we found Firm) was performed by an investigator blinded to the assigned that the activity of pyramidal neurons in the dorsal hippocam- treatments using a 2-chamber (15 cm × 15  cm × 20 cm) apparatus pus (dHC) after memory retrieval was critical for reconsolida- with distinct tactile environments to maximize contextual dif- tion. The results of RiboTag purification suggested that Tet3 ferences; a manual guillotine door (15 × 20  cm) separated the 2 transcripts were upregulated in CamkII neur α ons in the dHC chambers. On Day 1, mice were allowed to freely explore the after memory retrieval, and the total protein level of Tet3 in the entire apparatus for 15 minutes (pretest). The mice that stayed in dHC was elevated synchronously. Furthermore, knockdown of one chamber for more than 10 minutes were excluded from the Tet3 in pyramidal neurons in the dHC impaired the activation of experiment. On Days 2, 3, and 4, mice were given an i.p. injec- these neurons and memory reconsolidation. Our findings lead tion of cocaine (10 mg/kg) and confined to one of the chambers to the hypothesis that Tet3 in pyramidal neurons in the dHC (drug-paired) for 30 minutes; 6 hours later, they received an i.p. plays a pivotal role in reward memory reconsolidation and that Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 257 injection of saline (an equivalent volume to that of cocaine) and Electrophysiology were confined to the other chamber for 30 minutes (condition- Coronal sections (300  μm) containing the dHC were cut from mice ing). On Day 5, mice were allowed to freely explore the entire injected with AAV-CamkIIα-hM4D(i)-mCherry or AAV-CamkIIα- apparatus for 5 minutes (retrieval) and were randomized into 2 mCherry, and the slices were prepared as previously described groups, vehicle and Ani treatment, by tossing a coin. On Day 6, (Zhao et al., 2011). Briefly, the mice were anesthetized with iso- mice were allowed to freely explore the entire apparatus for 15 flurane and then transcardially perfused with cold protective minutes (test) to evaluate CPP scores. The CPP score (sec/min) was artificial cerebrospinal fluid (92  mM N-methyl-D-glucamine, defined as the time spent in the cocaine-paired chamber minus 2.5  mM KCl, 1.25  mM NaH PO , 30  mM NaHCO , 20  mM HEPES, 2 4 3 the time spent in the saline-paired chamber per minute. After 25 mM D-glucose, 2 mM thiourea, 5 mM Na ascorbate, 3 mM Na the experiments, all mice used in the cocaine-induced CPP were pyruvate, 0.5 mM CaCl, and 10 mM MgCl). After recovering at 2 2 sacrificed to verify virus expression and cannula positioning. 32°C to 34°C for 10 minutes, the slices were transferred into a holding chamber containing room-temperature carbogenated Virus Infection artificial cerebrospinal fluid (119 mM NaCl, 2.5 mM KCl, 1.25 mM The pSico vector (Addgene plasmid no. 11578) (Ventura et al., 2004) NaH PO , 26 mM NaHCO , 12.5 mM glucose, 2 mM CaCl, 2 mM 2 4 3 2 allows for stable, Cre-dependent expression of short hairpin RNAs MgCl , 2 mM thiourea, 5 mM Na ascorbate, and 3 mM Na pyru- (shRNAs) in cells and transgenic mice (Ventura et al., 2004). shRNAs vate) and stored for 45 minutes prior to recording. The slices encoding oligonucleotides (5’-GCTCCAACGAGAAGCTATTTG-3’) were used within 6 hours after preparation. targeting Tet3 (Ito et al., 2010) were subcloned into the pSico vec- Whole-cell current-clamp recordings were performed tor using Xhol/HpaI restriction sites. Lenti-Dio-Tet3 shRNA-EGFP from mCherry-positive and hM4D(i)-mCherry-positive neu- lentivirus was packaged by the Genechem Technology Co., Ltd., rons in the dHC with an EPC-10 amplifier and Patchmaster +8 +12 and the titer was 5E TU/mL. AAV-CamkIIα-Cre-EGFP (5E V.G./ software (HEKA Elektronik). The intracellular solution com- mL), AAV-CamkIIα-mCherry (5E V.G./mL), AAV-CamkIIα-hM4D(i)- position was as follows: 126  mM K-gluconate, 4  mM KCl, mCherry (5E V.G./mL), and AAV-RAM-d2tTA::TRE-FLEX-tdTomato 10  mM HEPES, 4  mM ATP-Mg, 0.3  mM GTP-Na , and 10  mM (5E V.G./mL) (Addgene plasmid no. 84468) (Sorensen et al., 2016) creatine phosphate (pH 7.2, 290–300 mOsm). The pipette were packaged by the Obio Technology Co, Ltd. Microinjections resistance was in the range of 8 to 10 Ω. Current-clamp were performed using 33-gauge injection needles connected to a recordings were filtered at 2.9 kHz and sampled at 5 kHz. To 10-μL Hamilton syringe. Each brain nucleus was injected with 0.2 confirm the inhibitory effect evoked by clozapine N-oxide μL of AAV combined with 1 μL of Lenti viruses at a slow injection (CNO), we injected 60 pA of current to elicit action potential rate (0.2 μL/min). The intended stereotactic coordinates were as firing. Current-clamp recording was performed before and follows: dorsal hippocampus –AP -1.5 mm; ML ±1.5 mm; D V -1.6 after bath application of 10  M CNO µ . mm; ACC – AP +0.5 mm; ML ±0.5 mm; DV -1.4 mm. All mice were given 2 to 3 weeks to recover before behavioral experiments, and RiboTag Purification the knockdown efficiency of the virus was checked by quantita- tive real time polymerase chain reaction (qRT-PCR). Purification of ribosome-associated mRNAs was conducted as described previously with slight modification (Sanz et al., 2009). Mice were decapitated, and the brains were removed immedi- ately. The dHCs were dissected by the coronal sections (according Immunohistochemistry to the stereotaxic coordinates from Bregma -1.0 mm to -2.5 mm) CamkIIα-cre::Rpl22-HA mice or virus-infected C57BL/6N mice were within 5 minutes in ice-cold PBS. The tissues were homogenized anesthetized with choral hydrate (400 mg/kg, i.p.), then perfused in 1 mL of supplemented hybridization buffer (25 mM Tris pH 7.0, transcardially with ice-cold saline and 4% paraformaldehyde (dis- 25 mM Tris pH 8.0, 12 mM MgCl, 100 mM KCl, 1% Triton X-100, solved in phosphate buffer [PB]). The brains were removed and 1 mM dithiothreitol (DTT), 1 pr × otease inhibitors [04693159001, postfixed with 4% PBS-paraformaldehyde solution overnight. Roche], 200 units/mL RNase inhibitor [N2112S, Promega], 100 Then the brains were subjected to dehydration in 20% sucrose μg/mL cycloheximide [14126, Cayman], and 1 mg/mL heparin) solution for 24 hours and 30% sucrose solution for 48 hours at using a tissue grinder (JXFSTPRP-24, Shanghai Jingxin Industrial 4ºC before being sliced into 30-μ m coronal sections. Slices were Development Co., Ltd). The supernatant was incubated with 10 washed in PBS and then incubated in primary antibody in block- μL of anti-HA antibody (H6908, Sigma) and 100 μ L of Dynabeads ing buffer (10% donkey serum in PBS containing 0.3% Triton X-100) Protein G (10003D, Novex, Invitrogen) for 12 hours. Purified mRNA overnight at 4ºC. The slices were washed and incubated in second- was eluted from the Dynabeads using TRIzol LS (Thermo Fisher ary antibody for 1 hour and 4 ,6-diamidino-2-phen ′ ylindole (DAPI) Scientific Inc) according to the manufacturer’s instructions with for 5 minutes at room temperature. After being washed in PBS, the inclusion of a DNase digestion step. An Agilent RNA 6000 Pico the slices were mounted in antiquenching mounting medium. For Kit (5067-1513, Agilent) and an Agilent 2100 bioanalyzer were 5-hydroxymethylcytosine (5hmC) staining, permeabilized cells used to evaluate the quality of purified mRNA. Samples with RIN were denatured as previously reported (Ito et al., 2010). The follow- numbers <7 were discarded. ing primary antibodies were used: anti-5hmC (4000, Active Motif, 1:500), anti-EAAC1 (AB1520, Millipore, 1:500), anti-GABA (A2052, RNA Extraction, Reverse Transcription, and Sigma, 1:500), anti-parvalbumin (AgPV100abs, Swant, 1:500), and Quantitative PCR Analyses anti-hemagglutinin (HA) antibodies (H6908-.5mL, Sigma, 1:1000). DAPI (D9564, Sigma, 1:100) was applied as a counterstain. The Total RNA was extracted from tissues using TRIzol Reagent following secondary antibodies were used: anti-mouse 488 (715- (Thermo Fisher Scientific Inc.) according to the manufactur - 545-150, Jackson ImmunoResearch, 1:500) and anti-rabbit 488 (711- er’s instructions. Reverse transcription was completed using a 545-152, Jackson ImmunoResearch, 1:500). Images were acquired PrimeScript RT Reagent Kit (RR037A, Takara). The cDNA was sub- using a Nikon-A1 confocal microscope with a 20 objecti × ve. jected to qRT-PCR using SYBR Premix Ex Taq (RR420A, Takara) Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 258 | International Journal of Neuropsychopharmacology, 2018 and an Eppendorf Mastercycler PCR System (Eppendorf). The in memory reconsolidation, we injected mice with AAV-CamkIIα- primers are listed in supplementary Table 1. mCherry or AAV-CamkIIα-hM4D(i)-mCherry in the dHC (dorsal CA1, dorsal CA3, and some dorsal regions of DG) (Figure  2a-b). Treatment with CNO, an activator of hM4D(i), repressed the fir - Western Blotting ing frequency of hM4D(i)-mCherry neurons (Figure 2c-d), while Brain tissues were lysed in RIPA buffer (50  mM Tris, pH 7.4, CNO had no effect on locomotion (supplementary Figure  1), 150  mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, confirming that the chemogenetic system works well.A AV- and protein inhibitors) and loaded onto Mini-PROTEAN TGX CamkIIα-mCherry or AAV-CamkIIα-hM4D(i)-mCherry was highly Precast Gels (Bio-Rad). Western blotting was carried out as expressed in dHC, but barely expressed in ventral hippocampus described before. Rabbit anti-Tet3 and anti-actin antibodies (Figure  2e). Two weeks after the viral infection, the mice were were purchased from GeneTex (GTX121453) and Sigma (A2066, trained to associate a context with cocaine (Figure  2a). During Darmstadt), respectively. IRDye 700CW-conjugated anti-rabbit the retrieval session, the 2 groups spent similar amounts of time antibody was purchased from Rockland Biosciences. The in- in the cocaine-paired chamber (Figure  2f, mCherry vs hM4D(i)- frared fluorescence images were obtained and quantified with mCherry, t = 1.032,P = .307). CNO (2 mg/kg, i.p.) was injected im- an Odyssey infrared imaging system (Li-Cor Bioscience). mediately after memory retrieval, and the CPP score was tested 24 hours later. The mice infected with AAV-CamkIIα-hM4D(i)- mCherry showed decreased CPP scores compared with the Statistical Analysis control group (Figure 2f, F (1, 16) = 4.937, P = .041, hM4D(i)- treatment Eight to 12 mice per group were used for behavioral tests, and mCherry vs mCherry, t = 2.39, P = .021), indicating that the activity 4 to 6 mice per group were used for qRT-PCR and biochemistry of pyramidal neurons in the dHC was required for cocaine-asso- studies. Data are presented as the mean ± SEM. The Student’s ciated memory reconsolidation. t test, 1-way ANOVA, or 2-way ANOVA was used for statistical analysis. Bonferroni posthoc analysis was performed after Exploring the Potential Regulators of Cocaine- 1-way or 2-way ANOVA. Significance levels were represented as Associated Memory Reconsolidation by Isolating *P < .05, **P < .01, and ***P < .001. Ribosome-Associated Transcripts from the Pyramidal Neurons in the dHC RESULTS The above data suggest that retrieval-induced expressional acti- vation of pyramidal neurons in the dHC plays a pivotal role in The Dorsal Hippocampus Is Required for Cocaine- memory reconsolidation of cocaine-induced CPP. Thus, we used Associated Memory Reconsolidation the RiboTag technique to isolate ribosome-associated (actively C57 mice were given an injection of cocaine (i.p., 10 mg/kg) and translated) mRNAs from the CamkIIα pyramidal neurons in trained daily for 3  days to associate the drug with one of the the dHC. CamkIIα-Cre::Rpl22-HA mice were generated by breed- CPP chambers. Twenty-four hours later, mice were reexposed ing CamkIIα-Cre mice with the RPL22-HA mouse line (Figure 3a), to the CPP chambers (retrieval), and they showed significant resulting in expression of HA-tagged ribosomes exclusively in preference for the cocaine-paired chamber, indicating that the CamkIIα neurons. Excitatory neuronal markers were enriched memory had been retrieved. Immediately after memory re- in RiboTag-purified transcripts of CamkIIα-Cre::Rpl22-HA mice, trieval, the protein biosynthesis inhibitor anisomycin (Ani, while inhibitory neuronal and glial markers were not (supple- 150 mg/kg, i.p.) or vehicle was injected i.p., and the mice were mentary Figure  2), indicating that the RiboTag purified excit- tested again 24 hours later for the retention (reconsolidation) atory-neuron transcripts. After cocaine-associated memory of cocaine-CPP memory (Figure 1a). The mice injected with Ani was formed (supplementary Figure 3) and retrieved (Figure 3b), after retrieval showed decreased time spent in the cocaine- ribosome-associated mRNAs in the dorsal hippocampi of paired chamber, indicating impaired reconsolidation (Figure 1b: CamkIIα-Cre::Rpl22-HA mice were isolated (Figure  3c), and the F (2, 44) = 114.439, P < .0001, F (2, 44) = 3.715, P = .0323, retrieval-induced changes in ribosome-associated mRNAs test treatment × test F (1, 22) = 9.576, P = .0053, Veh vs Ani within test, t = 4.060, (mRNAs that are being actively translated), especially those treatment P < .001). Both ACC and hippocampus play an important role related to neuronal activity, were assessed by qRT-PCR. As in contextual memory (Einarsson et  al., 2015). Furthermore, shown in Figure  3d to g, memory retrieval led to upregulation ACC to hippocampus (CA3 to CA1 region) projection mediates of the translational activity of the immediate early genes, such memory retrieval (Rajasethupathy et al., 2015), while their roles as Arc, Egr1, Fos, and Npas4 (Figure 3d, P = .0094, F(3, 19) = 5.373 for in memory reconsolidation are unclear. Bilateral infusion of Ani Arc; P < .0001, F(3, 19) = 15.66 for Egr1; P = .0072, F(3,19) = 5.763 for in the dHC (Figure  1c-d), but not in the ACC (Figure  1e-f), im- Fos; P = .0007, F(3, 19) = 9.560 for Npas4) (Bozon et al., 2003; Ploski mediately after the memory retrieval significantly inhibited the et al., 2011; Besnard et al., 2014; Lv et al., 2015; Webb et al., 2017). reconsolidation of cocaine CPP (Figure 1d: F (1, 17) = 6.486, Translation of mRNAs for neurotransmitters and receptors such treatment P = .014. Veh vs Ani within test, t = 2.999, P = .004; Figure 1f: F as Vgf, Trh, and Gpr68 were upregulated following retrieval, while treatment (1, 14) = 0.606, P = .449), indicating that the retrieval-induced syn- Scn2b was not (Figure 3e , P < .0001, F(3, 19) = 31.19 for Vgf; P = .0344, thesis of new protein in the dHC is necessary for the reconsoli- F(3, 19) = 3.683 for Trh; P = .0411, F(3, 19) = 3.469 for Gpr68; P = .0571, dation of cocaine-CPP memory. F(3, 19) = 3.085 for Scn2b). Previous studies have shown that the MAPK/ERK, PKA, and NFκB pathways are involved in memory reconsolidation (Kelly et  al., 2003Boccia et  ; al., 2007; Arguello The Postretrieval Activity of the Pyramidal Neurons et  al., 2014; de la Fuente et  al., 2015). However, the ribosome- in the dHC Is Required for the Reconsolidation of associated transcripts of several components of these signaling Cocaine-Associated Memory pathways, including Ikbkg, Prkcz, Pde7b, and Dusp1, had no signif- More than 90% of the neurons in the dHC are excitatory pyram- icant change after memory retrieval (Figure 3f). It is likely that the idal neurons. To test the potential role of the principle neurons activity but not the biosynthesis of those components induced Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 259 Figure 1. Inhibition of protein synthesis in the dorsal hippocampus disturbs reconsolidation of cocaine-associated memory. (a) Schematic diagram of experimental procedure. Mice were conditioned with cocaine in the conditioned place preference (CPP) apparatus for 3 days. One day after conditioning, mice were reexposed to the CPP apparatus (Retrieval), and the CPP score was determined. Immediately after the retrieval session, mice were given an i.p. injection of vehicle (Veh) or anisomycin (Ani), and CPP was tested 24 hours later (Test 2). (b) An i.p. injection of anisomycin significantly inhibited reconsolidation of cocaine-associated memor= y 11 for (n vehicle and n= 13 for anisomycin). (c,e) Schematic diagram of experimental procedure and cannula placement in the dorsal hippocampus (dHC) or anterior cingulate cortex (ACC). (d,f) Infusion of anisomycin into the dHC (d) but not the ACC (f) immediately after retrieval significantly inhibited cocaine-associated memory reconsoli- dation. dHC: n= 11 for vehicle and n= 8 for anisomycin. ACC: n= 9 for vehicle and n= 7 for anisomycin. **P< .01. Veh vs Ani. 2-way ANOVA. Data are presented as the mean ± SEM. by retrieval is essential for reconsolidation. Accumulating evi- P = .7460, F(3, 19) = 0.4129 for Hdac3), indicating the involvement dence suggests that the epigenetic machinery regulates the for - of epigenetic machinery in memory reconsolidation. mation and stabilization of long-term memory (Sultan and Day, 2011; Cassanelli et al., 2015). Intriguingly, the results of RiboTag Upregulation of Tet3 mRNA and Protein Level in the purification showed that ribosome-associated transcripts of dHC After Memory Retrieval epigenetic factors, including histone demethylase Kdm6b, his- tone methyltransferase, regulator of active DNA demethyla- Gadd45b plays an important role in active DNA demeth- tion Gadd45b, and histone deacetylase HDAC in the e 3 xcitatory ylation (Barreto et  al., 2007 Ma et  ; al., 2009; Jarome et  al., 2015) . neurons were upregulated in CamkIIα neurons after memory To explore the possibility that DNA demethylation takes part in retrieval (Figure 3gP , = .0045, F(3, 19) = 6.467 for Kdm6b; P = .0045, memory reconsolidation, we examined the level of 5-hydroxy- F(3, 19) = 6.458 for Cdyl; P = .0021, F(3, 19) = 7.506 for Gadd45b; methylcytosine (5hmC), a demethylation marker (Kaas et  al., Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 260 | International Journal of Neuropsychopharmacology, 2018 Figure  2. Activity of pyramidal neurons in the dorsal hippocampus is required for reconsolidation of cocaine-associated memory. (a–b) Schematic diagram (a) of chemo- genetic inhibition of pyramidal neurons in the dorsal hippocampus (dHC) (b) after retrieval. (c–d) Current-clamp recording of a representative mCherr neuron y and a hM4D(i)-mCherry neuron from cocaine-conditioned place preference (CPP)-trained mice. (c) Action potential firing before and after application of 10  µM CNO. Scale + + bar, 0.2  s, 20 mV. (d) Quantification of the firing frequency of mCherry neurons and hM4D(i)-mCherry neurons. n = 4 for each treatment; ** < P .01 Evoked vs Evoked + CNO. Paired Student’s test. t (e) Representative images of AAV-CamkII-mCherr α y-infected dHC and vHC slices. Viruses were barely expressed in ventral hippocampus. Red, mCherry; blue: DAPI. Scale bar, 2  mm. (f) Chemogenetic inhibition of the activity of pyramidal neurons in the dHC after retrieval impaired the reconsolidation of cocaine CPP. n = 8 for mCherry and n = 9 for hM4D(i)-mCherry; *P < .05 mCherry vs hM4D(i)-mCherry; 2-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. 2013), after memory retrieval. We observed a significant upregu- memory retrieval (Figure 4e , P = .0167, F(3, 21) = 4.275, P = .0129 for 0 lation of 5hmC level in pyramidal neurons of the dHC 4 hours minutes vs 1 hour; Figure 4f, P = .9880, F(3, 12) = 0.04154). after memory retrieval (Figure  4a-b , P = .0031, t = 6.377, unpaired Student’s t test). The ten-eleven translocation family of methyl- Tet3 in the Pyramidal Neurons of the dHC cytosine dioxygenases (including Tet1, Tet2, and Tet3), which Is Required for Cocaine-Associated Memory promote conversion from 5-methylcytosine to 5hmC, have been Reconsolidation proposed to play an important part in DNA demethylation regula- tion (Tahiliani et al., 2009Ito et  ; al., 2010; He et al., 2011). We found To examine the role of neuronal Tet3, we specifically knocked that although the total mRNA levels of Tet1 and Tet2 in the dHC down Tet3 in pyramidal neurons by co-infection of AAV-CamkIIα- were not significantly changed, the Tet3 mRNA increased after Cre-EGFP and Lenti-Dio-Tet3-shRNA-EGFP in the dHC of wild-type memory retrieval (Figure 4c P, = .0013, F(3, 17) = 8.253). Furthermore, mice that had acquired cocaine-induced CPP (Figure 5a-c). Tet3 ribosome-associated Tet3 mRNA in pyramidal neurons was signif- mRNA was significantly decreased in the dHC after lenti-Dio-Tet3 icantly upregulated after memory retrieval (Figure  P 4d = .0005, , F(3, shRNA-EGFP infection (Figure  5d, P = .00896, t = 6.074). The EGFP 16) = 10.22), while ribosome-associated Tet1 mRNA and Tet2 mRNA fluorescence was colocalized with glutamate transporter EAAC1, were not, indicating that Tet3 in pyramidal neurons of the dHC a marker of glutamatergic neurons (Lammel et al., 2012), not the might undergo active translation after memory retrieval. Western- interneuron marker PV or the inhibitory neuronal marker GABA, blotting analysis showed that the protein level of Tet3 in the dHC, indicating that these neurons in the dHC are glutamatergic but not in the ACC, was dramatically increased within 1 hour after neurons (supplementary Figure 4). During the memory retrieval Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 261 Figure 3. RiboTag purification of ribosome-associated mRNA in pyramidal neurons of the dorsal hippocampus with or without retrieval. (a) Breeding strategy for the CamkIIα-Cre::RPL22-HA (+/-) mice. (b) Experimental design for retrieval of cocaine-induced conditioned place preference (CPP) memory. (c) Scheme chart: dorsal hip- pocampi of CamkIIα-Cre::Rpl22-HA mice were separated. The dorsal hippocampi (dHCs) were dissected by the coronal sections (according to the stereotaxic coordinates from Bregma -1.0 mm to -2.5 mm) at 0 minutes (No Retrieval) (n = 6), 10 minutes (n= 4), 1 hour (n = 3), and 4 hours (n= 7) after retrieval of cocaine-induced CPP memory. RiboTag purification was conducted using anti-HA antibody and Protein G beads. Ribosome-associated mRNA in CamkII neur α ons was immunoprecipitated along with HA-tagged ribosomes. Purified mRNA was verified using qRT-PCR. (d–g) qRT-PCR analysis of ribosome-associated mRNAs of immediate early genes (d); neurotransmit- ters and receptors (e); critical molecules in the MAPK/ERK, PKA, and NFκB pathways (f); and epigenetic regulators (g) in pyramidal neurons at different time points after memory retrieval. ***P < .001, **P < .01, *P < .05. 1-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 262 | International Journal of Neuropsychopharmacology, 2018 Figure 4. Tet3 in the dorsal hippocampus (dHC) is upregulated after memory retrieval. (a–b) Genomic 5-hydroxymethylcytosine (5hmC) levels are significantly upregu- lated in the dHC of CamkIIα-Cre::RPL22-HA (+/-) mice after memory retrieval. Green, 5hmC; red, CamkIIα; blue, DAPI. Scale bar: upper, 100  µm; lower, 10 µm. n = 3 for No Retrieval and n= 3 for Retrieval. Data are presented as the mean ± SEM. **P < .01. Unpaired Student’ t test. s (c–d) qRT-PCR analysis of total (c: =n 6, 5, 5, and 5 for each group) or ribosome-associated (d: n = 6, 4, 3, and 7 for each group.) mRNA levels of Tet1, Tet2, and Tet3 in CamkII neur α ons from dorsal hippocampus at different time points after memory retrieval. *P < .05, **P < .01, *** <P .001 vs 0  hr (No Retrieval) Group. (e–f) Representative images of Western blots and quantification of the total Tet3 protein levels in the dHC (e: n = 7, 6, 6, and 6 for each group) and anterior cingulate cortex (ACC) (f: n = 4, 3, 3, and 3 for each group) at different time points after memory retrieval. *P < .05 vs 0 hr group. 1-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 263 Figure 5. Downregulation of Tet3 in pyramidal neurons of the dorsal hippocampus (dHC) impairs memory reconsolidation. (a) Schema chart. Cre-dependent expres- sion of EGFP/shRNA by a pSico lentivirus in CamkIIα neurons. (b) Experimental design. Downregulation of Tet3 in pyramidal neurons of the dHC after memory acqui- sition. (c) Representative images of virus expression in the dHC and anterior cingulate cortex (ACC). Green, EGFP; blue, DAPI. Scale bar µ, m. 40  (d) Verification of Tet3 knockdown efficiency. qRT-PCR of the Tet3 mRNA level in the dHC to verify the knockdown efficiency of the lentivirus Lenti-Dio-Tet3 shRNA-EGFP. Paired Student’s t test, n= 4. (e–f) Expression of Tet3 shRNA in pyramidal neurons in the dHC (e: n = 13 for scramble and n = 9 for Tet3 shRNA), but not in the ACC (f: n = 17 for scramble and n = 22 for Tet3 shRNA), after memory acquisition inhibited reconsolidation of cocaine-associated memory.< ** .01. P 2-way ANOVA with Bonferroni posthoc test. Data are presented as the mean ± SEM. (g–j) Schematic diagram of the viruses (g) and experimental procedure (h) used to label the activated pyramidal neurons in the dHC + - during memory retrieval. Quantification of the density of tdTomato EGFP cells in the dHC. Lenti-Dio-Tet3 shRNA-EGFP significantly downregulated the retrieval-induced activation of pyramidal neurons in the dHC (i,j). Red, tdTomato; Green, EGFP; Arrowheads, activated neurons, scale bar: 200  µm. n = 4 for scramble and n= 4 for Tet3 shRNA. Data are presented as the mean ± SEM. *P < .05. Unpaired Student’ ts test. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 264 | International Journal of Neuropsychopharmacology, 2018 session, the CPP score of mice co-injected with Lenti-Dio-Tet3- Epigenetic regulation plays a critical role in preserving long- shRNA-EGFP showed no difference compared with the Scramble term changes in neuronal cells. Maddox and colleagues (Maddox shRNA group, indicating that the ablation of Tet3 in pyramidal et  al., 2014) discovered that inhibiting DNA methyltransferase neurons does not affect memory retrieval (Figure  , F 5e (1, (DNMT) impairs fear memory reconsolidation. DNA demethyla- virus 80) = 4.044, P = .0477. Retrieval: scramble vs Tet3 shRNA, P = .6165). tion is typically thought to positively regulate gene transcription However, when assessed 24 hours after memory retrieval, the by promoting the binding of transcription factors (Levenson and CPP score of the Lenti-Dio-Tet3-shRNA-EGFP-infected group Sweatt, 2005; Levenson et al., 2006; Miller et al., 2010J; arome and was significantly decreased (Figure  5e, Test 3: scramble vs Tet3 Lubin, 2014), and we discovered that Tet3, which promotes DNA shRNA, P = .0014). In contrast, mice injected with Lenti-Dio-Tet3- demethylation, is critical for reward-related memory reconsoli- shRNA-EGFP in the ACC showed no change in CPP score during dation. It seems probable that both DNA demethylation and the retrieval and reconsolidation sessions (Figure  5f, F (1, methylation are required for memory updating and supported virus 148) = 1.454, P = .2298). These results identify an important and the notion that the dynamic procedures of epigenetic modifica- specific role of neuronal Tet3 in the dHC in regulating cocaine- tion, such as DNA demethylation and methylation, are import- associated memory reconsolidation. To test whether Tet3 ant for the formation of long-term memory. Our findings led regulates the activity of pyramidal neurons in the dHC, we co- to the hypothesis that DNA demethylation might take part in injected AAV-RAM-d2tTA::TRE-FLEX-tdTomato (Sorensen et  al., promoting the transcription of synaptic plasticity-related genes 2016) and Lenti-Dio-Tet3-shRNA-GFP into the dHC of CamkIIα-Cre during memory reconsolidation. mice after cocaine conditioning, then subjected these mice to The dynamic changes in Tet3 methylcytosine dioxygenases retrieval while off Dox (Figure  5g-h). Results showed that after after memory retrieval may be involved in retrieval-depend- + - memory retrieval, the density of tdTomato EGFP cells in the ent memory updating. Our findings raise the questions of dHC of Lenti-Dio-Tet3-shRNA-EGFP-infected group was significant what is the downstream target of Tet3 and how Tet3 regulates downregulated (Figure  5i-j, P = .0169, t = 3.276), indicating that downstream genes. Future studies aimed at target genes of Tet3 regulates the activity of pyramidal neurons in the dHC. Tet3 may contribute significantly to our understanding of the fundamental mechanism of the epigenetic machinery under - lying memory. Neuronal activation regulates the dynamics Discussion of the status of chromatin in a precisely timed manner, with Our study explored the critical molecules, cell type, and nuclei subsequent alterations in the gene expression profile. The identification of the components of different signaling path- required for cocaine-associated memory reconsolidation. We found that activity of excitatory pyramidal neurons in the dorsal ways involved in memory-related epigenetic regulation will provide mechanistic insights into the formation of long-term hippocampus is responsible for cocaine-associated memory re- consolidation. We also showed that memory retrieval led to memory. Of note is that all the memory experiments in this work were enhanced expression of Tet3, an important epigenetic modu- lator mediating DNA demethylation. Knockdown of Tet3 in conducted with the cocaine-induced CPP, which is a reward- associated memory model. Further studies using other mem- CamkIIα neurons of the dHC impaired the activation of pyram- idal neurons and cocaine-associated memory reconsolidation, ory paradigms such as auditory fear conditioning and inhibitory avoidance would help us better confirm the critical function of indicating that the upregulation of Tet3 is pivotal for memory reconsolidation. Tet3 in memory reconsolidation. In conclusion, we demonstrate for the first time, to our knowledge, that Tet3 of pyramidal neu- Given that memory reconsolidation is sensitive to protein synthesis inhibitors (Nader et al., 2000), it is reasonable to specu- rons in the dHC mediates cocaine-associated memory recon- solidation by acting as an epigenetic modulator. In particular, late that critical molecules mediating memory reconsolidation are among the proteins synthesized de novo upon memory our study illustrates that Tet3 may represent an exciting poten- tial target for the therapy of psychiatric disorders. retrieval. However, the study of cellular and molecular dynamics after memory retrieval is difficult: proteomics lacks cell specifi- city, while laser-capture microdissection and fluorescence-asso- Acknowledgments ciated cell sorting lack information on protein synthesis (Shin et  al., 2014). By dissecting target brain nuclei and applying This work was supported by grants from the Natural Science RiboTag purification (Heiman et al., 2008; Sanz et al., 2009), we Foundation of China (31430033, 31421091 and 91632307 to L.M. were able to circumvent previous limitations and capture ribo- and 31671042 to F.W.) and the Ministry of Science and Technology some-associated transcripts, which might be a partial, cell type- (2015CB553501 to L.M. and 2014CB942801 to L.M. and F.W.). specific representation of the proteins being synthesized. Based on our findings, ribosome-associated transcripts of immediate Statement of Interest early genes, ligands, and receptors were enhanced after mem- ory retrieval, which is in line with previous reports (Bozon et al., The authors declare no competing financial interests. 2003; Lv et al., 2015). However, no differences were detected in the molecules related to the MAPK/ERK pathway, PKA path- References way, and NFκB pathway, which are involved in memory recon- solidation (Kelly et  al., 2003Boccia et  ; al., 2007; Arguello et  al., Alberini CM, Kandel ER (2014) The regulation of transcription 2014). One explanation for this is that these signaling pathways in memory consolidation. Cold Spring Harb Perspect Biol might rely not only on translational activity but also, and to a 7:a021741. greater extent, on protein modification and interaction activity. Arguello AA, Hodges MA, Wells AM, Lara H 3rd, Xie X, Fuchs Accordingly, we confirmed the ribosome-associated transcripts RA (2014) Involvement of amygdalar protein kinase A, but of several well-known molecules involved in reconsolidation not calcium/calmodulin-dependent protein kinase II, in the were enhanced using RiboTag purification, indicating that our reconsolidation of cocaine-related contextual memories in system is reliable. rats. Psychopharmacology (Berl) 231:55–65. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 265 Barreto G, Schafer A, Marhold J, Stach D, Swaminathan SK, Jarome TJ, Butler AA, Nichols JN, Pacheco NL, Lubin FD (2015) Handa V, Doderlein G, Maltry N, Wu W, Lyko F, Niehrs C (2007) NF-kappaB mediates Gadd45beta expression and DNA dem- Gadd45a promotes epigenetic gene activation by repair- ethylation in the hippocampus during fear memory forma- mediated DNA demethylation. Nature 445:671–675. tion. Front Mol Neurosci 8:54. Besnard A, Laroche S, Caboche J (2014) Comparative dynamics Kaas GA, Zhong C, Eason DE, Ross DL, Vachhani RV, Ming GL, of MAPK/ERK signalling components and immediate early King JR, Song H, Sweatt JD (2013) TET1 controls CNS 5-methyl- genes in the hippocampus and amygdala following con- cytosine hydroxylation, active DNA demethylation, gene textual fear conditioning and retrieval. Brain Struct Funct transcription, and memory formation. Neuron 79:1086–1093. 219:415–430. Kelly A, Laroche S, Davis S (2003) Activation of mitogen-acti- Boccia M, Freudenthal R, Blake M, de la Fuente V, Acosta G, Baratti vated protein kinase/extracellular signal-regulated kinase in C, Romano A (2007) Activation of hippocampal nuclear factor- hippocampal circuitry is required for consolidation and re- kappa B by retrieval is required for memory reconsolidation. consolidation of recognition memory. J Neurosci 23:5354–5360. J Neurosci 27:13436–13445. Lammel S, Lim BK, Ran C, Huang KW, Betley MJ, Tye KM, Deisseroth Bozon B, Davis S, Laroche S (2003) A requirement for the imme- K, Malenka RC (2012) Input-specific control of reward and aver - diate early gene zif268 in reconsolidation of recognition sion in the ventral tegmental area. Nature 491:212–217. memory after retrieval. Neuron 40:695–701. Langemeijer SM, Aslanyan MG, Jansen JH (2009) TET proteins in Bredy TW, Barad M (2008) The histone deacetylase inhibitor val- malignant hematopoiesis. Cell Cycle 8:4044–4048. proic acid enhances acquisition, extinction, and reconsolida- Levenson JM, Sweatt JD (2005) Epigenetic mechanisms in mem- tion of conditioned fear. Learn Mem 15:39–45. ory formation. Nat Rev Neurosci 6:108–118. Cassanelli PM, Cladouchos ML, Fernandez Macedo G, Sifonios Levenson JM, Roth TL, Lubin FD, Miller CA, Huang IC, Desai P, L, Giaccardi LI, Gutierrez ML, Gravielle MC, Wikinski S (2015) Malone LM, Sweatt JD (2006) Evidence that DNA (cytosine-5) Working memory training triggers delayed chromatin methyltransferase regulates synaptic plasticity in the hippo- remodeling in the mouse corticostriatothalamic circuit. Prog campus. J Biol Chem 281:15763–15773. Neuropsychopharmacol Biol Psychiatry 60:93–103. Lv XF, Sun LL, Cui CL, Han JS (2015) NAc Shell Arc/Arg3.1 protein medi- Day JJ, Sweatt JD (2010) DNA methylation and memory forma- ates reconsolidation of morphine CPP by increased GluR1 cell tion. Nat Neurosci 13:1319–1323. surface expression: activation of ERK-coupled CREB is required. de la Fuente V, Federman N, Zalcman G, Salles A, Freudenthal R, Int J Neuropsychopharmacol 18: doi:10.1093/ijnp/pyv030. Romano A (2015) NF-kappaB transcription factor role in con- Ma DK, Jang MH, Guo JU, Kitabatake Y, Chang ML, Pow-Anpongkul solidation and reconsolidation of persistent memories. Front N, Flavell RA, Lu B, Ming GL, Song H (2009) Neuronal activity- Mol Neurosci 8:50. induced Gadd45b promotes epigenetic DNA demethylation Einarsson EO, Pors J, Nader K (2015) Systems reconsolida- and adult neurogenesis. Science 323:1074–1077. tion reveals a selective role for the anterior cingulate cor - Maddox SA, Watts CS, Schafe GE (2013) p300/CBP histone acetyl- tex in generalized contextual fear memory expression. transferase activity is required for newly acquired and reac- Neuropsychopharmacol 40:480–487. tivated fear memories in the lateral amygdala. Learn Mem Flavell CR, Lambert EA, Winters BD, Bredy TW (2013) Mechanisms 20:109–119. governing the reactivation-dependent destabilization of mem- Maddox SA, Watts CS, Schafe GE (2014) DNA methyltransferase ories and their role in extinction. Front Behav Neurosci 7:214. activity is required for memory-related neural plasticity in Graff J, Joseph NF, Horn ME, Samiei A, Meng J, Seo J, Rei D, Bero the lateral amygdala. Neurobiol Learn Memory 107:93–100. AW, Phan TX, Wagner F, Holson E, Xu J, Sun J, Neve RL, Mach Miller CA, Gavin CF, White JA, Parrish RR, Honasoge A, Yancey CR, RH, Haggarty SJ, Tsai LH (2014) Epigenetic priming of mem- Rivera IM, Rubio MD, Rumbaugh G, Sweatt JD (2010) Cortical ory updating during reconsolidation to attenuate remote fear DNA methylation maintains remote memory. Nat Neurosci memories. Cell 156:261–276. 13:664–666. Gu TP, Guo F, Yang H, Wu HP, Xu GF, Liu W, Xie ZG, Shi L, He X, Jin Nader K, Schafe GE, Le Doux JE (2000) Fear memories require SG, Iqbal K, Shi YG, Deng Z, Szabo PE, Pfeifer GP, Li J, Xu GL protein synthesis in the amygdala for reconsolidation after (2011) The role of Tet3 DNA dioxygenase in epigenetic repro- retrieval. Nature 406:722–726. gramming by oocytes. Nature 477:606–610. Oliveira AM (2016) DNA methylation: a permissive mark in Haubrich J, Nader K (2016) Memory reconsolidation. In: Current memory formation and maintenance. Learn Mem 23:587–593. topics in behavorial neurosiences, pp1–26. Berlin, Heidelberg: Ploski JE, Monsey MS, Nguyen T, DiLeone RJ, Schafe GE (2011) The Springer Berlin Heidelberg. neuronal PAS domain protein 4 (Npas4) is required for new He YF, Li BZ, Li Z, Liu P, Wang Y, Tang Q, Ding J, Jia Y, Chen Z, Li L, and reactivated fear memories. Plos One 6:e23760. Sun Y, Li X, Dai Q, Song CX, Zhang K, He C, Xu GL (2011) Tet- Rajasethupathy P, Sankaran S, Marshel JH, Kim CK, Ferenczi E, mediated formation of 5-carboxylcytosine and its excision by Lee SY, Berndt A, Ramakrishnan C, Jaffe A, Lo M, Liston C, TDG in mammalian DNA. Science 333:1303–1307. Deisseroth K (2015) Projections from neocortex mediate top- Heiman M, Schaefer A, Gong S, Peterson JD, Day M, Ramsey down control of memory retrieval. Nature 526:653–659. KE, Suarez-Farinas M, Schwarz C, Stephan DA, Surmeier Sanz E, Yang L, Su T, Morris DR, McKnight GS, Amieux PS DJ, Greengard P, Heintz N (2008) A translational profiling (2009) Cell-type-specific isolation of ribosome-associated approach for the molecular characterization of CNS cell mRNA from complex tissues. Proc Natl Acad Sci USA types. Cell 135:738–748. 106:13939–13944. Ito S, D’Alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y (2010) Shen L, Inoue A, He J, Liu Y, Lu F, Zhang Y (2014) Tet3 and DNA Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self- replication mediate demethylation of both the maternal and renewal and inner cell mass specification. Nature 466:1129–1133. paternal genomes in mouse zygotes. Cell Stem Cell 15:459–470. Jarome TJ, Lubin FD (2014) Epigenetic mechanisms of mem- Shin J, Ming GL, Song H (2014) Decoding neural transcriptomes ory formation and reconsolidation. Neurobiol Learn Mem and epigenomes via high-throughput sequencing. Nat 115:116–127. Neurosci 17:1463–1475. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 266 | International Journal of Neuropsychopharmacology, 2018 Sorensen AT, Cooper YA, Baratta MV, Weng FJ, Zhang Y, conditional RNA interference from transgenes. Proc Natl Ramamoorthi K, Fropf R, LaVerriere E, Xue J, Young A, Acad Sci USA 101:10380–10385. Schneider C, Gotzsche CR, Hemberg M, Yin JC, Maier SF, Lin Y Webb WM, Sanchez RG, Perez G, Butler AA, Hauser RM, Rich MC, (2016) A robust activity marking system for exploring active O’Bierne AL, Jarome TJ, Lubin FD (2017) Dynamic association neuronal ensembles. Elife 5 doi: 10.7554/eLife.13918. of epigenetic H3K4me3 and DNA 5hmC marks in the dorsal Sultan FA, Day JJ (2011) Epigenetic mechanisms in memory and hippocampus and anterior cingulate cortex following reacti- synaptic function. Epigenomics 3:157–181. vation of a fear memory. Neurobiol Learn Mem 142:66–78. Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Zhao S, Ting JT, Atallah HE, Qiu L, Tan J, Gloss B, Augustine GJ, Y, Agarwal S, Iyer LM, Liu DR, Aravind L, Rao A (2009) Deisseroth K, Luo M, Graybiel AM, Feng G (2011) Cell type- Conversion of 5-methylcytosine to 5-hydroxymethylcy- specific channelrhodopsin-2 transgenic mice for optogenetic tosine in mammalian DNA by MLL partner TET1. Science dissection of neural circuitry function. Nat Methods 8:745–752. 324:930–935. Zovkic IB, Guzman-Karlsson MC, Sweatt JD (2013) Epigenetic Ventura A, Meissner A, Dillon CP, McManus M, Sharp PA, regulation of memory formation and maintenance. Learn Van Parijs L, Jaenisch R, Jacks T (2004) Cre-lox-regulated Memory 20:61–74. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Neuropsychopharmacology Oxford University Press

Retrieval-Induced Upregulation of Tet3 in Pyramidal Neurons of the Dorsal Hippocampus Mediates Cocaine-Associated Memory Reconsolidation

Free
12 pages

Loading next page...
 
/lp/ou_press/retrieval-induced-upregulation-of-tet3-in-pyramidal-neurons-of-the-OrtystL8cs
Publisher
Collegium Internationale Neuro-Psychopharmacologicum
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of CINP.
ISSN
1461-1457
eISSN
1469-5111
D.O.I.
10.1093/ijnp/pyx099
Publisher site
See Article on Publisher Site

Abstract

Background: Memory retrieval refers to reexposure to information previously encoded and stored in the brain. Following retrieval, a once-consolidated memory destabilizes and undergoes reconsolidation, during which gene expression changes to restabilize memory. Investigating epigenetic regulation during reconsolidation could provide insights into normal memory formation and pathological memory associated with psychiatric disorders. Methods: We used cocaine-induced conditioned place preference to assess the cocaine-associated memory of mice and used chemogenetic methods to manipulate the activity of the pyramidal neurons in the dorsal hippocampus. We isolated the ribosome-associated transcripts from the excitatory neurons in the dorsal hippocampus by RiboTag purification to identify the potential epigenetic regulators, and we specifically knocked down gene expression in pyramidal neurons with a Cre- dependent lentivirus. Results: Chemogenetically silencing the activity of the pyramidal neurons in the dorsal hippocampus immediately after memory retrieval markedly impaired memory reconsolidation, and the ribosome-associated mRNA level of the ten-eleven translocation (Tet) family methylcytosine dioxygenase Tet3, but not Tet1 or Tet2, was dramatically upregulated 10 minutes after memory retrieval. The protein level of Tet3 in the dorsal hippocampus but not in the anterior cingulate cortex was dramatically increased 1 hour after memory retrieval. Specifically, knockdown of Tet3 in pyramidal neurons in the dorsal hippocampus decreased the activation of pyramidal neurons and impaired the reconsolidation of cocaine-associated memory. Conclusions: Our findings highlight the new function of the DNA demethylation regulator Tet3 in pyramidal neurons of the dorsal hippocampus in regulating the reconsolidation of cocaine-associated memory. Keywords: reconsolidation, dorsal hippocampus, CamkIIα neuron, Tet3, cocaine-associated memory Introduction The formation of long-term memory involves a series of protein synthesis, and neuronal plasticity dynamics (Flavell molecular and signaling changes, including gene transcription, et al., 2013). These changes may occur during learning and can Received: August 8, 2017; Revised: October 6, 2017; Accepted: October 27, 2017 © The Author(s) 2017. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, 255 provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 256 | International Journal of Neuropsychopharmacology, 2018 Significance Statement Immediately after memory retrieval, a once-consolidated memory destabilizes and undergoes reconsolidation, during which gene expression changes to restabilize memory. Investigating the epigenetic regulators that act during reconsolidation could provide insights into normal memory formation and pathological memory associated with psychiatric disorders. We found that the activity of the excitatory pyramidal neurons in the dorsal hippocampus (dHC) was required for cocaine-associated memory reconsolidation. After memory retrieval, the number of 5hmC pyramidal neurons and the ribosome-associated transcripts of the ten-eleven translocation (Tet) family methylcytosine dioxygenase Tet3, but not Tet1 or Tet2, were upregulated in the dHC. The protein level of Tet3 in the dHC was also increased after memory retrieval. Specifically, knockdown of Tet3 in pyramidal neurons of the dHC impaired the activation of pyramidal neurons and memory reconsolidation. Our findings indicate the importance of DNA demethylation of dHC in regulating reward memory reconsolidation and reveal Tet3 as a new regulator of reward memory. be subsequently retained for a long time. Memory reconsoli- Tet3 may represent a potential treatment target for memory- dation is the process in which previously consolidated memo- related disorders, including drug addiction. ries, having been reactivated, are stored again to maintain, strengthen, or modify existing memories (Haubrich and Nader, METHODS 2016). Following retrieval, once-consolidated memory destabi- lizes and requires gene expression to be restabilized (Jarome Animals and Housing and Lubin, 2014). The result of retrieval is a cascade of events T29-1Stl/J that ultimately leads to changes in neuronal plasticity, which CamkIIα-Cre (#005359 B6.Cg-Tg(Camk2a-cre) ), RPL22-HA tm1.1Psam/J mediate long-term memory updating (Alberini and Kandel, (#011029-B6N.129-Rpl22 ) mice were purchased from the 2014). Thus, understanding the molecular mechanisms of gene Jackson Laboratory. All mice were bred onto a C57BL/6J genetic transcription regulation during the reconsolidation processes background. We generated the CamkIIα-Cre::RPL22-HA (+/-) mice could provide crucial insights into normal memory formation by crossing CamkIIα-Cre mice with RPL22-HA mice. The 8- to and pathological memory associated with psychiatric disorders, 12-week-old male offspring were used in the experiments (see such as PTSD and addiction (Jarome and Lubin, 2014). Figure  3a for a breeding scheme). Genotypes were determined Memory retrieval induces de novo transcription and trans- by polymerase chain reaction (PCR) of mouse tail DNA samples. lation, which are orchestrated by epigenetic modifications and Mice were housed in groups on a 12-hour-light/-dark cycle with are pivotal for memory maintenance and modification (Alberini food and water available ad libitum. All animal treatment was and Kandel, 2014). Epigenetic mechanisms have been widely in strict accordance with the National Institutes of Health Guide implicated in synaptic plasticity underlying memory formation for the Care and Use of Laboratory Animals and was approved (Day and Sweatt, 2010; Sultan and Day, 2011; Zovkic et al., 2013). by the Animal Care and Use Committee of the School of Basic DNA methylation and histone modifications have emerged as Medical Sciences of Fudan University. critical transcriptional regulators of gene expression during ini- tial memory encoding (Jarome and Lubin, 2014), whereas studies Stereotaxic Surgery on epigenetic regulation in memory reconsolidation are few and mainly focused on histone acetylation (Bredy and Barad, 2008; Mice were anesthetized with choral hydrate (40 mg/kg, i.p.) and Maddox et  al., 2013; Graff et  al., 2014). Recent studies indicate placed in a stereotactic instrument. Guide cannulae (Plastics that DNA methylation is involved in the regulation of genomic One) were bilaterally implanted in the brain and affixed to responses during reconsolidation (Flavell et  al., 2013 Oli ; veira, the skull with dental cement. The intended stereotactic coor - 2016). By contrast, it remains unclear how memory retrieval dinates were as follows: dorsal hippocampus – anterior-poste- triggers epigenetic changes, especially demethylation changes, rior (AP) -1.5  mm; medial-lateral (ML) ±1.5  mm; dorsal-ventral during the reconsolidation phases to impact behavior, and new (DV) -1.5 mm; ACC – AP +0.5 mm; ML ±0.5 mm; DV -1.2 mm. All epigenetic regulators of gene transcription during memory re- mice were given at least 10  days to recover before behavioral consolidation remain to be identified. experiments. Anisomycin (A601115-0005, BBI) was dissolved Tet3 belongs to the ten-eleven translocation family of at 125 μg/μL as previously described (Nader et  al., 2000). Each methylcytosine dioxygenases, and it promotes DNA demethyla- mouse received a microinjection of 0.5 μL of anisomycin or vehi- tion (Langemeijer et al., 2009; Gu et al., 2011; Shen et al., 2014). cle through the cannula (0.2  μL/min) immediately after retrieval. Despite the abundance of Tet proteins in the brain, little is known about the dynamic changes and the functions of Tet enzymes Cocaine-Induced CPP after memory retrieval. In this study, by using a cocaine-induced CPP induced by cocaine hydrochloride (Qinghai Pharmaceutical conditioned place preference (CPP) paradigm in mice, we found Firm) was performed by an investigator blinded to the assigned that the activity of pyramidal neurons in the dorsal hippocam- treatments using a 2-chamber (15 cm × 15  cm × 20 cm) apparatus pus (dHC) after memory retrieval was critical for reconsolida- with distinct tactile environments to maximize contextual dif- tion. The results of RiboTag purification suggested that Tet3 ferences; a manual guillotine door (15 × 20  cm) separated the 2 transcripts were upregulated in CamkII neur α ons in the dHC chambers. On Day 1, mice were allowed to freely explore the after memory retrieval, and the total protein level of Tet3 in the entire apparatus for 15 minutes (pretest). The mice that stayed in dHC was elevated synchronously. Furthermore, knockdown of one chamber for more than 10 minutes were excluded from the Tet3 in pyramidal neurons in the dHC impaired the activation of experiment. On Days 2, 3, and 4, mice were given an i.p. injec- these neurons and memory reconsolidation. Our findings lead tion of cocaine (10 mg/kg) and confined to one of the chambers to the hypothesis that Tet3 in pyramidal neurons in the dHC (drug-paired) for 30 minutes; 6 hours later, they received an i.p. plays a pivotal role in reward memory reconsolidation and that Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 257 injection of saline (an equivalent volume to that of cocaine) and Electrophysiology were confined to the other chamber for 30 minutes (condition- Coronal sections (300  μm) containing the dHC were cut from mice ing). On Day 5, mice were allowed to freely explore the entire injected with AAV-CamkIIα-hM4D(i)-mCherry or AAV-CamkIIα- apparatus for 5 minutes (retrieval) and were randomized into 2 mCherry, and the slices were prepared as previously described groups, vehicle and Ani treatment, by tossing a coin. On Day 6, (Zhao et al., 2011). Briefly, the mice were anesthetized with iso- mice were allowed to freely explore the entire apparatus for 15 flurane and then transcardially perfused with cold protective minutes (test) to evaluate CPP scores. The CPP score (sec/min) was artificial cerebrospinal fluid (92  mM N-methyl-D-glucamine, defined as the time spent in the cocaine-paired chamber minus 2.5  mM KCl, 1.25  mM NaH PO , 30  mM NaHCO , 20  mM HEPES, 2 4 3 the time spent in the saline-paired chamber per minute. After 25 mM D-glucose, 2 mM thiourea, 5 mM Na ascorbate, 3 mM Na the experiments, all mice used in the cocaine-induced CPP were pyruvate, 0.5 mM CaCl, and 10 mM MgCl). After recovering at 2 2 sacrificed to verify virus expression and cannula positioning. 32°C to 34°C for 10 minutes, the slices were transferred into a holding chamber containing room-temperature carbogenated Virus Infection artificial cerebrospinal fluid (119 mM NaCl, 2.5 mM KCl, 1.25 mM The pSico vector (Addgene plasmid no. 11578) (Ventura et al., 2004) NaH PO , 26 mM NaHCO , 12.5 mM glucose, 2 mM CaCl, 2 mM 2 4 3 2 allows for stable, Cre-dependent expression of short hairpin RNAs MgCl , 2 mM thiourea, 5 mM Na ascorbate, and 3 mM Na pyru- (shRNAs) in cells and transgenic mice (Ventura et al., 2004). shRNAs vate) and stored for 45 minutes prior to recording. The slices encoding oligonucleotides (5’-GCTCCAACGAGAAGCTATTTG-3’) were used within 6 hours after preparation. targeting Tet3 (Ito et al., 2010) were subcloned into the pSico vec- Whole-cell current-clamp recordings were performed tor using Xhol/HpaI restriction sites. Lenti-Dio-Tet3 shRNA-EGFP from mCherry-positive and hM4D(i)-mCherry-positive neu- lentivirus was packaged by the Genechem Technology Co., Ltd., rons in the dHC with an EPC-10 amplifier and Patchmaster +8 +12 and the titer was 5E TU/mL. AAV-CamkIIα-Cre-EGFP (5E V.G./ software (HEKA Elektronik). The intracellular solution com- mL), AAV-CamkIIα-mCherry (5E V.G./mL), AAV-CamkIIα-hM4D(i)- position was as follows: 126  mM K-gluconate, 4  mM KCl, mCherry (5E V.G./mL), and AAV-RAM-d2tTA::TRE-FLEX-tdTomato 10  mM HEPES, 4  mM ATP-Mg, 0.3  mM GTP-Na , and 10  mM (5E V.G./mL) (Addgene plasmid no. 84468) (Sorensen et al., 2016) creatine phosphate (pH 7.2, 290–300 mOsm). The pipette were packaged by the Obio Technology Co, Ltd. Microinjections resistance was in the range of 8 to 10 Ω. Current-clamp were performed using 33-gauge injection needles connected to a recordings were filtered at 2.9 kHz and sampled at 5 kHz. To 10-μL Hamilton syringe. Each brain nucleus was injected with 0.2 confirm the inhibitory effect evoked by clozapine N-oxide μL of AAV combined with 1 μL of Lenti viruses at a slow injection (CNO), we injected 60 pA of current to elicit action potential rate (0.2 μL/min). The intended stereotactic coordinates were as firing. Current-clamp recording was performed before and follows: dorsal hippocampus –AP -1.5 mm; ML ±1.5 mm; D V -1.6 after bath application of 10  M CNO µ . mm; ACC – AP +0.5 mm; ML ±0.5 mm; DV -1.4 mm. All mice were given 2 to 3 weeks to recover before behavioral experiments, and RiboTag Purification the knockdown efficiency of the virus was checked by quantita- tive real time polymerase chain reaction (qRT-PCR). Purification of ribosome-associated mRNAs was conducted as described previously with slight modification (Sanz et al., 2009). Mice were decapitated, and the brains were removed immedi- ately. The dHCs were dissected by the coronal sections (according Immunohistochemistry to the stereotaxic coordinates from Bregma -1.0 mm to -2.5 mm) CamkIIα-cre::Rpl22-HA mice or virus-infected C57BL/6N mice were within 5 minutes in ice-cold PBS. The tissues were homogenized anesthetized with choral hydrate (400 mg/kg, i.p.), then perfused in 1 mL of supplemented hybridization buffer (25 mM Tris pH 7.0, transcardially with ice-cold saline and 4% paraformaldehyde (dis- 25 mM Tris pH 8.0, 12 mM MgCl, 100 mM KCl, 1% Triton X-100, solved in phosphate buffer [PB]). The brains were removed and 1 mM dithiothreitol (DTT), 1 pr × otease inhibitors [04693159001, postfixed with 4% PBS-paraformaldehyde solution overnight. Roche], 200 units/mL RNase inhibitor [N2112S, Promega], 100 Then the brains were subjected to dehydration in 20% sucrose μg/mL cycloheximide [14126, Cayman], and 1 mg/mL heparin) solution for 24 hours and 30% sucrose solution for 48 hours at using a tissue grinder (JXFSTPRP-24, Shanghai Jingxin Industrial 4ºC before being sliced into 30-μ m coronal sections. Slices were Development Co., Ltd). The supernatant was incubated with 10 washed in PBS and then incubated in primary antibody in block- μL of anti-HA antibody (H6908, Sigma) and 100 μ L of Dynabeads ing buffer (10% donkey serum in PBS containing 0.3% Triton X-100) Protein G (10003D, Novex, Invitrogen) for 12 hours. Purified mRNA overnight at 4ºC. The slices were washed and incubated in second- was eluted from the Dynabeads using TRIzol LS (Thermo Fisher ary antibody for 1 hour and 4 ,6-diamidino-2-phen ′ ylindole (DAPI) Scientific Inc) according to the manufacturer’s instructions with for 5 minutes at room temperature. After being washed in PBS, the inclusion of a DNase digestion step. An Agilent RNA 6000 Pico the slices were mounted in antiquenching mounting medium. For Kit (5067-1513, Agilent) and an Agilent 2100 bioanalyzer were 5-hydroxymethylcytosine (5hmC) staining, permeabilized cells used to evaluate the quality of purified mRNA. Samples with RIN were denatured as previously reported (Ito et al., 2010). The follow- numbers <7 were discarded. ing primary antibodies were used: anti-5hmC (4000, Active Motif, 1:500), anti-EAAC1 (AB1520, Millipore, 1:500), anti-GABA (A2052, RNA Extraction, Reverse Transcription, and Sigma, 1:500), anti-parvalbumin (AgPV100abs, Swant, 1:500), and Quantitative PCR Analyses anti-hemagglutinin (HA) antibodies (H6908-.5mL, Sigma, 1:1000). DAPI (D9564, Sigma, 1:100) was applied as a counterstain. The Total RNA was extracted from tissues using TRIzol Reagent following secondary antibodies were used: anti-mouse 488 (715- (Thermo Fisher Scientific Inc.) according to the manufactur - 545-150, Jackson ImmunoResearch, 1:500) and anti-rabbit 488 (711- er’s instructions. Reverse transcription was completed using a 545-152, Jackson ImmunoResearch, 1:500). Images were acquired PrimeScript RT Reagent Kit (RR037A, Takara). The cDNA was sub- using a Nikon-A1 confocal microscope with a 20 objecti × ve. jected to qRT-PCR using SYBR Premix Ex Taq (RR420A, Takara) Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 258 | International Journal of Neuropsychopharmacology, 2018 and an Eppendorf Mastercycler PCR System (Eppendorf). The in memory reconsolidation, we injected mice with AAV-CamkIIα- primers are listed in supplementary Table 1. mCherry or AAV-CamkIIα-hM4D(i)-mCherry in the dHC (dorsal CA1, dorsal CA3, and some dorsal regions of DG) (Figure  2a-b). Treatment with CNO, an activator of hM4D(i), repressed the fir - Western Blotting ing frequency of hM4D(i)-mCherry neurons (Figure 2c-d), while Brain tissues were lysed in RIPA buffer (50  mM Tris, pH 7.4, CNO had no effect on locomotion (supplementary Figure  1), 150  mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, confirming that the chemogenetic system works well.A AV- and protein inhibitors) and loaded onto Mini-PROTEAN TGX CamkIIα-mCherry or AAV-CamkIIα-hM4D(i)-mCherry was highly Precast Gels (Bio-Rad). Western blotting was carried out as expressed in dHC, but barely expressed in ventral hippocampus described before. Rabbit anti-Tet3 and anti-actin antibodies (Figure  2e). Two weeks after the viral infection, the mice were were purchased from GeneTex (GTX121453) and Sigma (A2066, trained to associate a context with cocaine (Figure  2a). During Darmstadt), respectively. IRDye 700CW-conjugated anti-rabbit the retrieval session, the 2 groups spent similar amounts of time antibody was purchased from Rockland Biosciences. The in- in the cocaine-paired chamber (Figure  2f, mCherry vs hM4D(i)- frared fluorescence images were obtained and quantified with mCherry, t = 1.032,P = .307). CNO (2 mg/kg, i.p.) was injected im- an Odyssey infrared imaging system (Li-Cor Bioscience). mediately after memory retrieval, and the CPP score was tested 24 hours later. The mice infected with AAV-CamkIIα-hM4D(i)- mCherry showed decreased CPP scores compared with the Statistical Analysis control group (Figure 2f, F (1, 16) = 4.937, P = .041, hM4D(i)- treatment Eight to 12 mice per group were used for behavioral tests, and mCherry vs mCherry, t = 2.39, P = .021), indicating that the activity 4 to 6 mice per group were used for qRT-PCR and biochemistry of pyramidal neurons in the dHC was required for cocaine-asso- studies. Data are presented as the mean ± SEM. The Student’s ciated memory reconsolidation. t test, 1-way ANOVA, or 2-way ANOVA was used for statistical analysis. Bonferroni posthoc analysis was performed after Exploring the Potential Regulators of Cocaine- 1-way or 2-way ANOVA. Significance levels were represented as Associated Memory Reconsolidation by Isolating *P < .05, **P < .01, and ***P < .001. Ribosome-Associated Transcripts from the Pyramidal Neurons in the dHC RESULTS The above data suggest that retrieval-induced expressional acti- vation of pyramidal neurons in the dHC plays a pivotal role in The Dorsal Hippocampus Is Required for Cocaine- memory reconsolidation of cocaine-induced CPP. Thus, we used Associated Memory Reconsolidation the RiboTag technique to isolate ribosome-associated (actively C57 mice were given an injection of cocaine (i.p., 10 mg/kg) and translated) mRNAs from the CamkIIα pyramidal neurons in trained daily for 3  days to associate the drug with one of the the dHC. CamkIIα-Cre::Rpl22-HA mice were generated by breed- CPP chambers. Twenty-four hours later, mice were reexposed ing CamkIIα-Cre mice with the RPL22-HA mouse line (Figure 3a), to the CPP chambers (retrieval), and they showed significant resulting in expression of HA-tagged ribosomes exclusively in preference for the cocaine-paired chamber, indicating that the CamkIIα neurons. Excitatory neuronal markers were enriched memory had been retrieved. Immediately after memory re- in RiboTag-purified transcripts of CamkIIα-Cre::Rpl22-HA mice, trieval, the protein biosynthesis inhibitor anisomycin (Ani, while inhibitory neuronal and glial markers were not (supple- 150 mg/kg, i.p.) or vehicle was injected i.p., and the mice were mentary Figure  2), indicating that the RiboTag purified excit- tested again 24 hours later for the retention (reconsolidation) atory-neuron transcripts. After cocaine-associated memory of cocaine-CPP memory (Figure 1a). The mice injected with Ani was formed (supplementary Figure 3) and retrieved (Figure 3b), after retrieval showed decreased time spent in the cocaine- ribosome-associated mRNAs in the dorsal hippocampi of paired chamber, indicating impaired reconsolidation (Figure 1b: CamkIIα-Cre::Rpl22-HA mice were isolated (Figure  3c), and the F (2, 44) = 114.439, P < .0001, F (2, 44) = 3.715, P = .0323, retrieval-induced changes in ribosome-associated mRNAs test treatment × test F (1, 22) = 9.576, P = .0053, Veh vs Ani within test, t = 4.060, (mRNAs that are being actively translated), especially those treatment P < .001). Both ACC and hippocampus play an important role related to neuronal activity, were assessed by qRT-PCR. As in contextual memory (Einarsson et  al., 2015). Furthermore, shown in Figure  3d to g, memory retrieval led to upregulation ACC to hippocampus (CA3 to CA1 region) projection mediates of the translational activity of the immediate early genes, such memory retrieval (Rajasethupathy et al., 2015), while their roles as Arc, Egr1, Fos, and Npas4 (Figure 3d, P = .0094, F(3, 19) = 5.373 for in memory reconsolidation are unclear. Bilateral infusion of Ani Arc; P < .0001, F(3, 19) = 15.66 for Egr1; P = .0072, F(3,19) = 5.763 for in the dHC (Figure  1c-d), but not in the ACC (Figure  1e-f), im- Fos; P = .0007, F(3, 19) = 9.560 for Npas4) (Bozon et al., 2003; Ploski mediately after the memory retrieval significantly inhibited the et al., 2011; Besnard et al., 2014; Lv et al., 2015; Webb et al., 2017). reconsolidation of cocaine CPP (Figure 1d: F (1, 17) = 6.486, Translation of mRNAs for neurotransmitters and receptors such treatment P = .014. Veh vs Ani within test, t = 2.999, P = .004; Figure 1f: F as Vgf, Trh, and Gpr68 were upregulated following retrieval, while treatment (1, 14) = 0.606, P = .449), indicating that the retrieval-induced syn- Scn2b was not (Figure 3e , P < .0001, F(3, 19) = 31.19 for Vgf; P = .0344, thesis of new protein in the dHC is necessary for the reconsoli- F(3, 19) = 3.683 for Trh; P = .0411, F(3, 19) = 3.469 for Gpr68; P = .0571, dation of cocaine-CPP memory. F(3, 19) = 3.085 for Scn2b). Previous studies have shown that the MAPK/ERK, PKA, and NFκB pathways are involved in memory reconsolidation (Kelly et  al., 2003Boccia et  ; al., 2007; Arguello The Postretrieval Activity of the Pyramidal Neurons et  al., 2014; de la Fuente et  al., 2015). However, the ribosome- in the dHC Is Required for the Reconsolidation of associated transcripts of several components of these signaling Cocaine-Associated Memory pathways, including Ikbkg, Prkcz, Pde7b, and Dusp1, had no signif- More than 90% of the neurons in the dHC are excitatory pyram- icant change after memory retrieval (Figure 3f). It is likely that the idal neurons. To test the potential role of the principle neurons activity but not the biosynthesis of those components induced Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 259 Figure 1. Inhibition of protein synthesis in the dorsal hippocampus disturbs reconsolidation of cocaine-associated memory. (a) Schematic diagram of experimental procedure. Mice were conditioned with cocaine in the conditioned place preference (CPP) apparatus for 3 days. One day after conditioning, mice were reexposed to the CPP apparatus (Retrieval), and the CPP score was determined. Immediately after the retrieval session, mice were given an i.p. injection of vehicle (Veh) or anisomycin (Ani), and CPP was tested 24 hours later (Test 2). (b) An i.p. injection of anisomycin significantly inhibited reconsolidation of cocaine-associated memor= y 11 for (n vehicle and n= 13 for anisomycin). (c,e) Schematic diagram of experimental procedure and cannula placement in the dorsal hippocampus (dHC) or anterior cingulate cortex (ACC). (d,f) Infusion of anisomycin into the dHC (d) but not the ACC (f) immediately after retrieval significantly inhibited cocaine-associated memory reconsoli- dation. dHC: n= 11 for vehicle and n= 8 for anisomycin. ACC: n= 9 for vehicle and n= 7 for anisomycin. **P< .01. Veh vs Ani. 2-way ANOVA. Data are presented as the mean ± SEM. by retrieval is essential for reconsolidation. Accumulating evi- P = .7460, F(3, 19) = 0.4129 for Hdac3), indicating the involvement dence suggests that the epigenetic machinery regulates the for - of epigenetic machinery in memory reconsolidation. mation and stabilization of long-term memory (Sultan and Day, 2011; Cassanelli et al., 2015). Intriguingly, the results of RiboTag Upregulation of Tet3 mRNA and Protein Level in the purification showed that ribosome-associated transcripts of dHC After Memory Retrieval epigenetic factors, including histone demethylase Kdm6b, his- tone methyltransferase, regulator of active DNA demethyla- Gadd45b plays an important role in active DNA demeth- tion Gadd45b, and histone deacetylase HDAC in the e 3 xcitatory ylation (Barreto et  al., 2007 Ma et  ; al., 2009; Jarome et  al., 2015) . neurons were upregulated in CamkIIα neurons after memory To explore the possibility that DNA demethylation takes part in retrieval (Figure 3gP , = .0045, F(3, 19) = 6.467 for Kdm6b; P = .0045, memory reconsolidation, we examined the level of 5-hydroxy- F(3, 19) = 6.458 for Cdyl; P = .0021, F(3, 19) = 7.506 for Gadd45b; methylcytosine (5hmC), a demethylation marker (Kaas et  al., Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 260 | International Journal of Neuropsychopharmacology, 2018 Figure  2. Activity of pyramidal neurons in the dorsal hippocampus is required for reconsolidation of cocaine-associated memory. (a–b) Schematic diagram (a) of chemo- genetic inhibition of pyramidal neurons in the dorsal hippocampus (dHC) (b) after retrieval. (c–d) Current-clamp recording of a representative mCherr neuron y and a hM4D(i)-mCherry neuron from cocaine-conditioned place preference (CPP)-trained mice. (c) Action potential firing before and after application of 10  µM CNO. Scale + + bar, 0.2  s, 20 mV. (d) Quantification of the firing frequency of mCherry neurons and hM4D(i)-mCherry neurons. n = 4 for each treatment; ** < P .01 Evoked vs Evoked + CNO. Paired Student’s test. t (e) Representative images of AAV-CamkII-mCherr α y-infected dHC and vHC slices. Viruses were barely expressed in ventral hippocampus. Red, mCherry; blue: DAPI. Scale bar, 2  mm. (f) Chemogenetic inhibition of the activity of pyramidal neurons in the dHC after retrieval impaired the reconsolidation of cocaine CPP. n = 8 for mCherry and n = 9 for hM4D(i)-mCherry; *P < .05 mCherry vs hM4D(i)-mCherry; 2-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. 2013), after memory retrieval. We observed a significant upregu- memory retrieval (Figure 4e , P = .0167, F(3, 21) = 4.275, P = .0129 for 0 lation of 5hmC level in pyramidal neurons of the dHC 4 hours minutes vs 1 hour; Figure 4f, P = .9880, F(3, 12) = 0.04154). after memory retrieval (Figure  4a-b , P = .0031, t = 6.377, unpaired Student’s t test). The ten-eleven translocation family of methyl- Tet3 in the Pyramidal Neurons of the dHC cytosine dioxygenases (including Tet1, Tet2, and Tet3), which Is Required for Cocaine-Associated Memory promote conversion from 5-methylcytosine to 5hmC, have been Reconsolidation proposed to play an important part in DNA demethylation regula- tion (Tahiliani et al., 2009Ito et  ; al., 2010; He et al., 2011). We found To examine the role of neuronal Tet3, we specifically knocked that although the total mRNA levels of Tet1 and Tet2 in the dHC down Tet3 in pyramidal neurons by co-infection of AAV-CamkIIα- were not significantly changed, the Tet3 mRNA increased after Cre-EGFP and Lenti-Dio-Tet3-shRNA-EGFP in the dHC of wild-type memory retrieval (Figure 4c P, = .0013, F(3, 17) = 8.253). Furthermore, mice that had acquired cocaine-induced CPP (Figure 5a-c). Tet3 ribosome-associated Tet3 mRNA in pyramidal neurons was signif- mRNA was significantly decreased in the dHC after lenti-Dio-Tet3 icantly upregulated after memory retrieval (Figure  P 4d = .0005, , F(3, shRNA-EGFP infection (Figure  5d, P = .00896, t = 6.074). The EGFP 16) = 10.22), while ribosome-associated Tet1 mRNA and Tet2 mRNA fluorescence was colocalized with glutamate transporter EAAC1, were not, indicating that Tet3 in pyramidal neurons of the dHC a marker of glutamatergic neurons (Lammel et al., 2012), not the might undergo active translation after memory retrieval. Western- interneuron marker PV or the inhibitory neuronal marker GABA, blotting analysis showed that the protein level of Tet3 in the dHC, indicating that these neurons in the dHC are glutamatergic but not in the ACC, was dramatically increased within 1 hour after neurons (supplementary Figure 4). During the memory retrieval Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 261 Figure 3. RiboTag purification of ribosome-associated mRNA in pyramidal neurons of the dorsal hippocampus with or without retrieval. (a) Breeding strategy for the CamkIIα-Cre::RPL22-HA (+/-) mice. (b) Experimental design for retrieval of cocaine-induced conditioned place preference (CPP) memory. (c) Scheme chart: dorsal hip- pocampi of CamkIIα-Cre::Rpl22-HA mice were separated. The dorsal hippocampi (dHCs) were dissected by the coronal sections (according to the stereotaxic coordinates from Bregma -1.0 mm to -2.5 mm) at 0 minutes (No Retrieval) (n = 6), 10 minutes (n= 4), 1 hour (n = 3), and 4 hours (n= 7) after retrieval of cocaine-induced CPP memory. RiboTag purification was conducted using anti-HA antibody and Protein G beads. Ribosome-associated mRNA in CamkII neur α ons was immunoprecipitated along with HA-tagged ribosomes. Purified mRNA was verified using qRT-PCR. (d–g) qRT-PCR analysis of ribosome-associated mRNAs of immediate early genes (d); neurotransmit- ters and receptors (e); critical molecules in the MAPK/ERK, PKA, and NFκB pathways (f); and epigenetic regulators (g) in pyramidal neurons at different time points after memory retrieval. ***P < .001, **P < .01, *P < .05. 1-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 262 | International Journal of Neuropsychopharmacology, 2018 Figure 4. Tet3 in the dorsal hippocampus (dHC) is upregulated after memory retrieval. (a–b) Genomic 5-hydroxymethylcytosine (5hmC) levels are significantly upregu- lated in the dHC of CamkIIα-Cre::RPL22-HA (+/-) mice after memory retrieval. Green, 5hmC; red, CamkIIα; blue, DAPI. Scale bar: upper, 100  µm; lower, 10 µm. n = 3 for No Retrieval and n= 3 for Retrieval. Data are presented as the mean ± SEM. **P < .01. Unpaired Student’ t test. s (c–d) qRT-PCR analysis of total (c: =n 6, 5, 5, and 5 for each group) or ribosome-associated (d: n = 6, 4, 3, and 7 for each group.) mRNA levels of Tet1, Tet2, and Tet3 in CamkII neur α ons from dorsal hippocampus at different time points after memory retrieval. *P < .05, **P < .01, *** <P .001 vs 0  hr (No Retrieval) Group. (e–f) Representative images of Western blots and quantification of the total Tet3 protein levels in the dHC (e: n = 7, 6, 6, and 6 for each group) and anterior cingulate cortex (ACC) (f: n = 4, 3, 3, and 3 for each group) at different time points after memory retrieval. *P < .05 vs 0 hr group. 1-way ANOVA with Bonferroni’s posthoc test. Data are presented as the mean ± SEM. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 263 Figure 5. Downregulation of Tet3 in pyramidal neurons of the dorsal hippocampus (dHC) impairs memory reconsolidation. (a) Schema chart. Cre-dependent expres- sion of EGFP/shRNA by a pSico lentivirus in CamkIIα neurons. (b) Experimental design. Downregulation of Tet3 in pyramidal neurons of the dHC after memory acqui- sition. (c) Representative images of virus expression in the dHC and anterior cingulate cortex (ACC). Green, EGFP; blue, DAPI. Scale bar µ, m. 40  (d) Verification of Tet3 knockdown efficiency. qRT-PCR of the Tet3 mRNA level in the dHC to verify the knockdown efficiency of the lentivirus Lenti-Dio-Tet3 shRNA-EGFP. Paired Student’s t test, n= 4. (e–f) Expression of Tet3 shRNA in pyramidal neurons in the dHC (e: n = 13 for scramble and n = 9 for Tet3 shRNA), but not in the ACC (f: n = 17 for scramble and n = 22 for Tet3 shRNA), after memory acquisition inhibited reconsolidation of cocaine-associated memory.< ** .01. P 2-way ANOVA with Bonferroni posthoc test. Data are presented as the mean ± SEM. (g–j) Schematic diagram of the viruses (g) and experimental procedure (h) used to label the activated pyramidal neurons in the dHC + - during memory retrieval. Quantification of the density of tdTomato EGFP cells in the dHC. Lenti-Dio-Tet3 shRNA-EGFP significantly downregulated the retrieval-induced activation of pyramidal neurons in the dHC (i,j). Red, tdTomato; Green, EGFP; Arrowheads, activated neurons, scale bar: 200  µm. n = 4 for scramble and n= 4 for Tet3 shRNA. Data are presented as the mean ± SEM. *P < .05. Unpaired Student’ ts test. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 264 | International Journal of Neuropsychopharmacology, 2018 session, the CPP score of mice co-injected with Lenti-Dio-Tet3- Epigenetic regulation plays a critical role in preserving long- shRNA-EGFP showed no difference compared with the Scramble term changes in neuronal cells. Maddox and colleagues (Maddox shRNA group, indicating that the ablation of Tet3 in pyramidal et  al., 2014) discovered that inhibiting DNA methyltransferase neurons does not affect memory retrieval (Figure  , F 5e (1, (DNMT) impairs fear memory reconsolidation. DNA demethyla- virus 80) = 4.044, P = .0477. Retrieval: scramble vs Tet3 shRNA, P = .6165). tion is typically thought to positively regulate gene transcription However, when assessed 24 hours after memory retrieval, the by promoting the binding of transcription factors (Levenson and CPP score of the Lenti-Dio-Tet3-shRNA-EGFP-infected group Sweatt, 2005; Levenson et al., 2006; Miller et al., 2010J; arome and was significantly decreased (Figure  5e, Test 3: scramble vs Tet3 Lubin, 2014), and we discovered that Tet3, which promotes DNA shRNA, P = .0014). In contrast, mice injected with Lenti-Dio-Tet3- demethylation, is critical for reward-related memory reconsoli- shRNA-EGFP in the ACC showed no change in CPP score during dation. It seems probable that both DNA demethylation and the retrieval and reconsolidation sessions (Figure  5f, F (1, methylation are required for memory updating and supported virus 148) = 1.454, P = .2298). These results identify an important and the notion that the dynamic procedures of epigenetic modifica- specific role of neuronal Tet3 in the dHC in regulating cocaine- tion, such as DNA demethylation and methylation, are import- associated memory reconsolidation. To test whether Tet3 ant for the formation of long-term memory. Our findings led regulates the activity of pyramidal neurons in the dHC, we co- to the hypothesis that DNA demethylation might take part in injected AAV-RAM-d2tTA::TRE-FLEX-tdTomato (Sorensen et  al., promoting the transcription of synaptic plasticity-related genes 2016) and Lenti-Dio-Tet3-shRNA-GFP into the dHC of CamkIIα-Cre during memory reconsolidation. mice after cocaine conditioning, then subjected these mice to The dynamic changes in Tet3 methylcytosine dioxygenases retrieval while off Dox (Figure  5g-h). Results showed that after after memory retrieval may be involved in retrieval-depend- + - memory retrieval, the density of tdTomato EGFP cells in the ent memory updating. Our findings raise the questions of dHC of Lenti-Dio-Tet3-shRNA-EGFP-infected group was significant what is the downstream target of Tet3 and how Tet3 regulates downregulated (Figure  5i-j, P = .0169, t = 3.276), indicating that downstream genes. Future studies aimed at target genes of Tet3 regulates the activity of pyramidal neurons in the dHC. Tet3 may contribute significantly to our understanding of the fundamental mechanism of the epigenetic machinery under - lying memory. Neuronal activation regulates the dynamics Discussion of the status of chromatin in a precisely timed manner, with Our study explored the critical molecules, cell type, and nuclei subsequent alterations in the gene expression profile. The identification of the components of different signaling path- required for cocaine-associated memory reconsolidation. We found that activity of excitatory pyramidal neurons in the dorsal ways involved in memory-related epigenetic regulation will provide mechanistic insights into the formation of long-term hippocampus is responsible for cocaine-associated memory re- consolidation. We also showed that memory retrieval led to memory. Of note is that all the memory experiments in this work were enhanced expression of Tet3, an important epigenetic modu- lator mediating DNA demethylation. Knockdown of Tet3 in conducted with the cocaine-induced CPP, which is a reward- associated memory model. Further studies using other mem- CamkIIα neurons of the dHC impaired the activation of pyram- idal neurons and cocaine-associated memory reconsolidation, ory paradigms such as auditory fear conditioning and inhibitory avoidance would help us better confirm the critical function of indicating that the upregulation of Tet3 is pivotal for memory reconsolidation. Tet3 in memory reconsolidation. In conclusion, we demonstrate for the first time, to our knowledge, that Tet3 of pyramidal neu- Given that memory reconsolidation is sensitive to protein synthesis inhibitors (Nader et al., 2000), it is reasonable to specu- rons in the dHC mediates cocaine-associated memory recon- solidation by acting as an epigenetic modulator. In particular, late that critical molecules mediating memory reconsolidation are among the proteins synthesized de novo upon memory our study illustrates that Tet3 may represent an exciting poten- tial target for the therapy of psychiatric disorders. retrieval. However, the study of cellular and molecular dynamics after memory retrieval is difficult: proteomics lacks cell specifi- city, while laser-capture microdissection and fluorescence-asso- Acknowledgments ciated cell sorting lack information on protein synthesis (Shin et  al., 2014). By dissecting target brain nuclei and applying This work was supported by grants from the Natural Science RiboTag purification (Heiman et al., 2008; Sanz et al., 2009), we Foundation of China (31430033, 31421091 and 91632307 to L.M. were able to circumvent previous limitations and capture ribo- and 31671042 to F.W.) and the Ministry of Science and Technology some-associated transcripts, which might be a partial, cell type- (2015CB553501 to L.M. and 2014CB942801 to L.M. and F.W.). specific representation of the proteins being synthesized. Based on our findings, ribosome-associated transcripts of immediate Statement of Interest early genes, ligands, and receptors were enhanced after mem- ory retrieval, which is in line with previous reports (Bozon et al., The authors declare no competing financial interests. 2003; Lv et al., 2015). However, no differences were detected in the molecules related to the MAPK/ERK pathway, PKA path- References way, and NFκB pathway, which are involved in memory recon- solidation (Kelly et  al., 2003Boccia et  ; al., 2007; Arguello et  al., Alberini CM, Kandel ER (2014) The regulation of transcription 2014). One explanation for this is that these signaling pathways in memory consolidation. Cold Spring Harb Perspect Biol might rely not only on translational activity but also, and to a 7:a021741. greater extent, on protein modification and interaction activity. Arguello AA, Hodges MA, Wells AM, Lara H 3rd, Xie X, Fuchs Accordingly, we confirmed the ribosome-associated transcripts RA (2014) Involvement of amygdalar protein kinase A, but of several well-known molecules involved in reconsolidation not calcium/calmodulin-dependent protein kinase II, in the were enhanced using RiboTag purification, indicating that our reconsolidation of cocaine-related contextual memories in system is reliable. rats. Psychopharmacology (Berl) 231:55–65. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Liu et al. | 265 Barreto G, Schafer A, Marhold J, Stach D, Swaminathan SK, Jarome TJ, Butler AA, Nichols JN, Pacheco NL, Lubin FD (2015) Handa V, Doderlein G, Maltry N, Wu W, Lyko F, Niehrs C (2007) NF-kappaB mediates Gadd45beta expression and DNA dem- Gadd45a promotes epigenetic gene activation by repair- ethylation in the hippocampus during fear memory forma- mediated DNA demethylation. Nature 445:671–675. tion. Front Mol Neurosci 8:54. Besnard A, Laroche S, Caboche J (2014) Comparative dynamics Kaas GA, Zhong C, Eason DE, Ross DL, Vachhani RV, Ming GL, of MAPK/ERK signalling components and immediate early King JR, Song H, Sweatt JD (2013) TET1 controls CNS 5-methyl- genes in the hippocampus and amygdala following con- cytosine hydroxylation, active DNA demethylation, gene textual fear conditioning and retrieval. Brain Struct Funct transcription, and memory formation. Neuron 79:1086–1093. 219:415–430. Kelly A, Laroche S, Davis S (2003) Activation of mitogen-acti- Boccia M, Freudenthal R, Blake M, de la Fuente V, Acosta G, Baratti vated protein kinase/extracellular signal-regulated kinase in C, Romano A (2007) Activation of hippocampal nuclear factor- hippocampal circuitry is required for consolidation and re- kappa B by retrieval is required for memory reconsolidation. consolidation of recognition memory. J Neurosci 23:5354–5360. J Neurosci 27:13436–13445. Lammel S, Lim BK, Ran C, Huang KW, Betley MJ, Tye KM, Deisseroth Bozon B, Davis S, Laroche S (2003) A requirement for the imme- K, Malenka RC (2012) Input-specific control of reward and aver - diate early gene zif268 in reconsolidation of recognition sion in the ventral tegmental area. Nature 491:212–217. memory after retrieval. Neuron 40:695–701. Langemeijer SM, Aslanyan MG, Jansen JH (2009) TET proteins in Bredy TW, Barad M (2008) The histone deacetylase inhibitor val- malignant hematopoiesis. Cell Cycle 8:4044–4048. proic acid enhances acquisition, extinction, and reconsolida- Levenson JM, Sweatt JD (2005) Epigenetic mechanisms in mem- tion of conditioned fear. Learn Mem 15:39–45. ory formation. Nat Rev Neurosci 6:108–118. Cassanelli PM, Cladouchos ML, Fernandez Macedo G, Sifonios Levenson JM, Roth TL, Lubin FD, Miller CA, Huang IC, Desai P, L, Giaccardi LI, Gutierrez ML, Gravielle MC, Wikinski S (2015) Malone LM, Sweatt JD (2006) Evidence that DNA (cytosine-5) Working memory training triggers delayed chromatin methyltransferase regulates synaptic plasticity in the hippo- remodeling in the mouse corticostriatothalamic circuit. Prog campus. J Biol Chem 281:15763–15773. Neuropsychopharmacol Biol Psychiatry 60:93–103. Lv XF, Sun LL, Cui CL, Han JS (2015) NAc Shell Arc/Arg3.1 protein medi- Day JJ, Sweatt JD (2010) DNA methylation and memory forma- ates reconsolidation of morphine CPP by increased GluR1 cell tion. Nat Neurosci 13:1319–1323. surface expression: activation of ERK-coupled CREB is required. de la Fuente V, Federman N, Zalcman G, Salles A, Freudenthal R, Int J Neuropsychopharmacol 18: doi:10.1093/ijnp/pyv030. Romano A (2015) NF-kappaB transcription factor role in con- Ma DK, Jang MH, Guo JU, Kitabatake Y, Chang ML, Pow-Anpongkul solidation and reconsolidation of persistent memories. Front N, Flavell RA, Lu B, Ming GL, Song H (2009) Neuronal activity- Mol Neurosci 8:50. induced Gadd45b promotes epigenetic DNA demethylation Einarsson EO, Pors J, Nader K (2015) Systems reconsolida- and adult neurogenesis. Science 323:1074–1077. tion reveals a selective role for the anterior cingulate cor - Maddox SA, Watts CS, Schafe GE (2013) p300/CBP histone acetyl- tex in generalized contextual fear memory expression. transferase activity is required for newly acquired and reac- Neuropsychopharmacol 40:480–487. tivated fear memories in the lateral amygdala. Learn Mem Flavell CR, Lambert EA, Winters BD, Bredy TW (2013) Mechanisms 20:109–119. governing the reactivation-dependent destabilization of mem- Maddox SA, Watts CS, Schafe GE (2014) DNA methyltransferase ories and their role in extinction. Front Behav Neurosci 7:214. activity is required for memory-related neural plasticity in Graff J, Joseph NF, Horn ME, Samiei A, Meng J, Seo J, Rei D, Bero the lateral amygdala. Neurobiol Learn Memory 107:93–100. AW, Phan TX, Wagner F, Holson E, Xu J, Sun J, Neve RL, Mach Miller CA, Gavin CF, White JA, Parrish RR, Honasoge A, Yancey CR, RH, Haggarty SJ, Tsai LH (2014) Epigenetic priming of mem- Rivera IM, Rubio MD, Rumbaugh G, Sweatt JD (2010) Cortical ory updating during reconsolidation to attenuate remote fear DNA methylation maintains remote memory. Nat Neurosci memories. Cell 156:261–276. 13:664–666. Gu TP, Guo F, Yang H, Wu HP, Xu GF, Liu W, Xie ZG, Shi L, He X, Jin Nader K, Schafe GE, Le Doux JE (2000) Fear memories require SG, Iqbal K, Shi YG, Deng Z, Szabo PE, Pfeifer GP, Li J, Xu GL protein synthesis in the amygdala for reconsolidation after (2011) The role of Tet3 DNA dioxygenase in epigenetic repro- retrieval. Nature 406:722–726. gramming by oocytes. Nature 477:606–610. Oliveira AM (2016) DNA methylation: a permissive mark in Haubrich J, Nader K (2016) Memory reconsolidation. In: Current memory formation and maintenance. Learn Mem 23:587–593. topics in behavorial neurosiences, pp1–26. Berlin, Heidelberg: Ploski JE, Monsey MS, Nguyen T, DiLeone RJ, Schafe GE (2011) The Springer Berlin Heidelberg. neuronal PAS domain protein 4 (Npas4) is required for new He YF, Li BZ, Li Z, Liu P, Wang Y, Tang Q, Ding J, Jia Y, Chen Z, Li L, and reactivated fear memories. Plos One 6:e23760. Sun Y, Li X, Dai Q, Song CX, Zhang K, He C, Xu GL (2011) Tet- Rajasethupathy P, Sankaran S, Marshel JH, Kim CK, Ferenczi E, mediated formation of 5-carboxylcytosine and its excision by Lee SY, Berndt A, Ramakrishnan C, Jaffe A, Lo M, Liston C, TDG in mammalian DNA. Science 333:1303–1307. Deisseroth K (2015) Projections from neocortex mediate top- Heiman M, Schaefer A, Gong S, Peterson JD, Day M, Ramsey down control of memory retrieval. Nature 526:653–659. KE, Suarez-Farinas M, Schwarz C, Stephan DA, Surmeier Sanz E, Yang L, Su T, Morris DR, McKnight GS, Amieux PS DJ, Greengard P, Heintz N (2008) A translational profiling (2009) Cell-type-specific isolation of ribosome-associated approach for the molecular characterization of CNS cell mRNA from complex tissues. Proc Natl Acad Sci USA types. Cell 135:738–748. 106:13939–13944. Ito S, D’Alessio AC, Taranova OV, Hong K, Sowers LC, Zhang Y (2010) Shen L, Inoue A, He J, Liu Y, Lu F, Zhang Y (2014) Tet3 and DNA Role of Tet proteins in 5mC to 5hmC conversion, ES-cell self- replication mediate demethylation of both the maternal and renewal and inner cell mass specification. Nature 466:1129–1133. paternal genomes in mouse zygotes. Cell Stem Cell 15:459–470. Jarome TJ, Lubin FD (2014) Epigenetic mechanisms of mem- Shin J, Ming GL, Song H (2014) Decoding neural transcriptomes ory formation and reconsolidation. Neurobiol Learn Mem and epigenomes via high-throughput sequencing. Nat 115:116–127. Neurosci 17:1463–1475. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018 266 | International Journal of Neuropsychopharmacology, 2018 Sorensen AT, Cooper YA, Baratta MV, Weng FJ, Zhang Y, conditional RNA interference from transgenes. Proc Natl Ramamoorthi K, Fropf R, LaVerriere E, Xue J, Young A, Acad Sci USA 101:10380–10385. Schneider C, Gotzsche CR, Hemberg M, Yin JC, Maier SF, Lin Y Webb WM, Sanchez RG, Perez G, Butler AA, Hauser RM, Rich MC, (2016) A robust activity marking system for exploring active O’Bierne AL, Jarome TJ, Lubin FD (2017) Dynamic association neuronal ensembles. Elife 5 doi: 10.7554/eLife.13918. of epigenetic H3K4me3 and DNA 5hmC marks in the dorsal Sultan FA, Day JJ (2011) Epigenetic mechanisms in memory and hippocampus and anterior cingulate cortex following reacti- synaptic function. Epigenomics 3:157–181. vation of a fear memory. Neurobiol Learn Mem 142:66–78. Tahiliani M, Koh KP, Shen Y, Pastor WA, Bandukwala H, Brudno Zhao S, Ting JT, Atallah HE, Qiu L, Tan J, Gloss B, Augustine GJ, Y, Agarwal S, Iyer LM, Liu DR, Aravind L, Rao A (2009) Deisseroth K, Luo M, Graybiel AM, Feng G (2011) Cell type- Conversion of 5-methylcytosine to 5-hydroxymethylcy- specific channelrhodopsin-2 transgenic mice for optogenetic tosine in mammalian DNA by MLL partner TET1. Science dissection of neural circuitry function. Nat Methods 8:745–752. 324:930–935. Zovkic IB, Guzman-Karlsson MC, Sweatt JD (2013) Epigenetic Ventura A, Meissner A, Dillon CP, McManus M, Sharp PA, regulation of memory formation and maintenance. Learn Van Parijs L, Jaenisch R, Jacks T (2004) Cre-lox-regulated Memory 20:61–74. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/3/255/4587482 by Ed 'DeepDyve' Gillespie user on 16 March 2018

Journal

International Journal of NeuropsychopharmacologyOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off